Open Nava2 opened 6 months ago
The filterAdvice
task generates a report at reports/dependency-analysis/final-advice.json
, and this report is machine-readable. Is this not sufficient? The plugin also supports users adding their own post-processing task which will automatically ingest that report and give you access to the full data model of the advice. See the wiki.
I think the JSON format is much better than what I currently have been using. However, it would still be nice to have sarif
format or another widely used format built-in to avoid building them from scratch. 👍🏻
Thanks for the issue. I do not have time to implement this request. I imagine there must be many competing "standards", and I can't implement them all. Maybe what you want is some kind of service loader framework and a plugin system to auto-convert the plugin's canonical output to something custom.
Describe the solution you'd like New output format for
projectHealth
andbuildHealth
.In order to support features like Github Checks, we need the ability to see line-numbers and other metadata associated with files specifically.
In order to reduce the feature request, if we generated
errorformat
outputs, we could hook into tools like @reviewdog/reviewdog.Describe alternatives you've considered
I wrote up a github action with a python script to parse the
buildHealth.txt
file for outputs: https://github.com/Nava2/dependency-analysis-reviewdog-action. This works OK, but it is brittle and prone to errors as the formats change.Another, better example than
errorformat
is probably usingsarif
(Link)